As I mentioned a few weeks ago, part of the reason my postings have been limited recently is due to my focus on PyVote development. This development has been in support of a series of experiments I have been conducting to research the effect of time constraints on legislative dynamics. Regular readers may remember that the original motivation for starting PyVote was to try to understand how democratic legislatures react to terrorist events—specifically—what types of legislation do democracies generate in the wake of these events.
My hypothesis for why we observe the types of laws we do is that those who control the legislative agenda take advantage of the perceived time constraint by proposing bills that mirror their personal preference, rather than that of the legislative body as a whole. To test this, I have used PyVote to conduct a series of experiments I call the “greedy conservative” simulations. A detailed description of the relevant theory, motivation, experimental design, and results is now available for download as a working paper entitled, “Evidence for Enhanced Agenda-Setter Control Under Legislative Time Constraint.”
Abstract: Extreme external events can propel legislature to quickly create new laws that address these unforeseen circumstances. In these cases, however, it is often observed that the resulting legislation does not reflect the majority ideologies of the given legislative body, but rather the ideological extremes. A possible reason for these outcomes is an enhanced level of control afforded to the agenda-setter given the legislative time constraint. To test this hypothesis, an agent-based model of a legislature is constructed, where committees are formed to create new bills, and agent utility is negatively affected based on the amount of time agents take to pass a bill. The experiment is called the “greedy conservative” simulation, wherein the agenda-setter on each committee is always the most conservative member, and that agent always proposes a bill at exactly its ideal point in the first voting period. The results of these experiments provide substantial evidence for enhanced agenda-setter control in the presence of a legislative time constraint, illuminating an aspect of the underlying legislative dynamics that contribute to laws being passed that are odds with many legislators’ ideological preferences.
The code for the experiments is available at the PyVote website, and I am eager to get reader feedback on the theory, design and results. Particularly, what’s wrong with it and where should this research go next? I conclude the paper with some ideas for future research, but I am interested in any of your ideas.
Automatically Generated Related posts:
Popular Posts